--- Day changed Sat Apr 18 2009 | ||
-!- DrTread [n=chatzill@cpe-70-112-223-97.austin.res.rr.com] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032608]"] | 00:00 | |
-!- myelinza1 [i=bryan@dhcp-84-36.me.utexas.edu] has quit [Read error: 60 (Operation timed out)] | 00:01 | |
DrTread_ | ok, this is fun. good night, world. we shall most likely conquer you in the morning. | 00:02 |
---|---|---|
-!- DrTread_ [n=irchon@cpe-70-112-223-97.austin.res.rr.com] has quit [Remote closed the connection] | 00:02 | |
-!- myelinzar [i=bryan@dhcp-84-36.me.utexas.edu] has joined #hplusroadmap | 00:06 | |
-!- genehacker [n=chatzill@wireless-128-62-96-159.public.utexas.edu] has joined #hplusroadmap | 00:08 | |
genehacker | fenn where can I download that sketchflat program? | 00:10 |
-!- PeerInfinity [n=someone@216.36.180.162] has quit [] | 00:48 | |
-!- any29470031 is now known as katsmeow-afk | 00:58 | |
genehacker | I don't believe it | 01:05 |
genehacker | I just do not believe it | 01:05 |
genehacker | http://groups.google.com/group/sci.nanotech/msg/96a67c84809c9a5d | 01:15 |
genehacker | kanzure, I do believe this belongs in your archives | 01:15 |
genehacker | I heard about fluidic replicators, but I never read the full proposal | 01:16 |
genehacker | and guess what? | 01:16 |
genehacker | a 8086 fluidic based processor with reasonably sized logic gates should be about a cubic foot | 01:20 |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has joined #hplusroadmap | 01:30 | |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has quit [] | 02:04 | |
kanzure | uugghhh | 02:12 |
kanzure | this is terrible | 02:12 |
kanzure | okay, they just don't know how to turn off the building | 02:17 |
kanzure | genehacker: yeah, chris phoenix from CRN | 02:17 |
genehacker | what is? | 02:18 |
genehacker | turn off building? | 02:19 |
genehacker | turn off fluid flow | 02:19 |
kanzure | mitch porter. yeah, I've emailed mitch before. | 02:21 |
kanzure | the building is speaking to me | 02:21 |
kanzure | it thinks its on fire | 02:21 |
kanzure | it now has a thousand angry college-aged men and women very angry, and inside of it | 02:21 |
kanzure | ready to attack. | 02:21 |
genehacker | fuck? | 02:22 |
genehacker | the castillon is? | 02:22 |
genehacker | what the hell is going on there? | 02:23 |
genehacker | fire alarm going off? | 02:23 |
genehacker | if you smell smoke GTFO | 02:23 |
genehacker | well I hope you GTFO ok if everything is fine | 02:24 |
genehacker | would you like me to carry out visual aquisition of the building to guage the situation? | 02:25 |
kanzure | no, I confirmed that they are bullshitting by asking them. | 02:26 |
genehacker | what the hell is going on down there? | 02:27 |
kanzure | I think they have some newbies at the front desk | 02:28 |
kanzure | who don't know how to shut off the alarms | 02:28 |
kanzure | or, they possibly fell asleep | 02:28 |
genehacker | I heard that happens often down there is that true? | 02:30 |
kanzure | not really. | 02:30 |
kanzure | genehacker: will you post that link to OM? | 02:30 |
genehacker | sure just not right now | 02:30 |
kanzure | ok I will then | 02:30 |
genehacker | if you want to me to do so I will | 02:30 |
kanzure | yes | 02:31 |
kanzure | forget it | 02:31 |
kanzure | yay it stopped | 02:32 |
kanzure | >30 min. jeebus. | 02:32 |
genehacker | posted | 02:33 |
kanzure | "the building emergency condition has been cleared. you may return to your normal activity." <- x5. | 02:34 |
kanzure | now in spanish. | 02:34 |
kanzure | ugh | 02:34 |
genehacker | I had something like that happen in highschool | 02:34 |
* kanzure stabs someone | 02:34 | |
genehacker | during spainish class | 02:34 |
genehacker | damn | 02:35 |
genehacker | that was fast kanzure | 02:35 |
genehacker | so I did some oft napkin calculations | 02:36 |
genehacker | we could make a 8086 level processor with shrinky dink microfluidics | 02:37 |
genehacker | that would take up about a 2.5 cm cube | 02:37 |
fenn | that link is stupid | 02:54 |
fenn | he glosses over the whole materials problem with magic super UV-cure plastic | 02:54 |
kanzure | yes | 03:01 |
kanzure | drtread should be able to fix that | 03:01 |
kanzure | or at least yell at him about it | 03:01 |
-!- splicer [n=patrik@h8n3c1o261.bredband.skanova.com] has joined #hplusroadmap | 03:12 | |
genehacker | wait a second | 03:14 |
genehacker | sounds like he's using arms | 03:14 |
genehacker | 7-dof arms | 03:14 |
genehacker | ugh | 03:15 |
-!- nsh [n=nsh@host86-133-35-168.range86-133.btcentralplus.com] has quit [Remote closed the connection] | 03:24 | |
-!- genehacker [n=chatzill@wireless-128-62-96-159.public.utexas.edu] has quit [Read error: 110 (Connection timed out)] | 03:54 | |
-!- Netsplit wolfe.freenode.net <-> irc.freenode.net quits: truename | 04:02 | |
-!- Netsplit over, joins: truename | 04:03 | |
-!- amaruk [n=amaruk@p4FDC14FA.dip0.t-ipconnect.de] has joined #hplusroadmap | 04:45 | |
-!- amaruk [n=amaruk@p4FDC14FA.dip0.t-ipconnect.de] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032608]"] | 05:08 | |
-!- any81332573 [n=someone@75-121-60-222.dyn.centurytel.net] has joined #hplusroadmap | 08:54 | |
-!- katsmeow-afk [n=someone@75-120-16-254.dyn.centurytel.net] has quit [Read error: 113 (No route to host)] | 09:06 | |
-!- jm [n=jm@p57B9C51E.dip.t-dialin.net] has joined #hplusroadmap | 09:20 | |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has joined #hplusroadmap | 10:26 | |
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)] | 10:54 | |
-!- jm|space [n=jm@p57B9D33D.dip.t-dialin.net] has joined #hplusroadmap | 11:10 | |
-!- jm [n=jm@p57B9C51E.dip.t-dialin.net] has quit [Success] | 11:26 | |
-!- any81332573 is now known as katsmeow-afk | 12:03 | |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has quit [Read error: 104 (Connection reset by peer)] | 12:07 | |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has joined #hplusroadmap | 12:07 | |
kanzure | fenn, what if a shape doesn't fall into a predefined category? | 12:16 |
-!- xp_prg [n=xp_prg3@208-106-52-115.adsl.dynamic.shocking.com] has joined #hplusroadmap | 12:25 | |
-!- PeerInfinity [n=someone@216.36.180.162] has joined #hplusroadmap | 12:54 | |
-!- cis-action [n=cis-acti@206.111.159.66.ptr.us.xo.net] has quit [] | 13:02 | |
-!- samrose [n=samrose@ip-207-145-38-45.iad.megapath.net] has joined #hplusroadmap | 13:19 | |
fenn | then approximate it | 14:04 |
fenn | or evaluate it exactly if you can figure out how | 14:05 |
kanzure | what do you mean by approximate | 14:05 |
fenn | like that catia gear example | 14:05 |
kanzure | right | 14:05 |
kanzure | so I was thinking of two types of shapes for this purpose | 14:05 |
kanzure | (1) tagged shape | 14:05 |
kanzure | (2) exact CAD data | 14:05 |
kanzure | if there's enough repeats of type #2s being used, | 14:05 |
kanzure | then it's time to write a quick script to replace all instances with a tag | 14:05 |
kanzure | so as to compress the information | 14:05 |
fenn | approximate like a bunch of line segments | 14:06 |
fenn | (not really, since you've got all sorts of curves at your perusal) | 14:06 |
fenn | and modeling something like cloth is always going to be hard, since it's stochastic | 14:07 |
kanzure | did you look at mtt yet? | 14:14 |
fenn | it's over my head | 14:17 |
kanzure | wonder how much of it is obfuscation because an academician wrote it, versus actually being complicated | 14:18 |
fenn | honestly it sounds like ADL bullshit | 14:18 |
fenn | look, a graph! your problems are solved. | 14:18 |
kanzure | it implements some sort of port/interface-based-physical-quantities-algebra-thingy | 14:19 |
kanzure | anywho. in python, I was going to make it so that the 'magnitude range' of physical quantities is defined by boolean algebraic expressions, i.e. typical if(condition1 && condition2 && (blah || blah2)) stuff | 14:20 |
kanzure | should I just store if(stuff) the stuff variable somewhere? | 14:20 |
kanzure | I mean, I want to define 'stuff' as a string | 14:20 |
kanzure | but it's equivalent to a conditional expression that needs to get evaluated at some point in the future | 14:20 |
kanzure | ideally I would like to avoid eval() | 14:21 |
fenn | yes, avoid eval() at all costs | 14:21 |
fenn | you shouldn't have to use it at all; if you are, you're coding perl or tcl or some other language | 14:21 |
kanzure | is it sick that I used to have a CMS project going where the pages were all eval()'d? :( | 14:21 |
fenn | put all the logic in an evaluation function | 14:21 |
kanzure | in a what? | 14:22 |
fenn | a function, that is called | 14:22 |
fenn | to evaluate the constraint | 14:22 |
kanzure | I guess I can have a class that represents a hierchical binary logic tree thingy | 14:22 |
kanzure | and evaluates the tree | 14:22 |
fenn | i still dont know how to make use of knowledge of the constraint | 14:22 |
kanzure | the operators are: less than, greater than, equal to, less than or equal to, greater than or equal to, plus or minus | 14:22 |
fenn | like, a human could look at the constraints and analytically determine some of the proper values | 14:23 |
fenn | but some iterative constraint solver is basically blind | 14:23 |
kanzure | this is not a constraint solver. | 14:23 |
kanzure | it's a thingy saying "hey look the ranges don't match up!" | 14:23 |
fenn | eh? | 14:23 |
fenn | oh, well, same thing | 14:23 |
kanzure | no? | 14:24 |
fenn | evaluator is part of the solver | 14:24 |
kanzure | a constraint solver takes on some form of a multivariable function optimization thingy | 14:24 |
kanzure | this is not what this is. | 14:24 |
fenn | while !solved(): fiddle_values() | 14:24 |
kanzure | what does that have to do with the evaluation of a boolean logic expression to represent an adequate 'range' of values | 14:25 |
-!- genehacker [n=chatzill@wireless-128-62-173-3.public.utexas.edu] has joined #hplusroadmap | 14:25 | |
kanzure | thus my list of operators above | 14:25 |
fenn | you stuff the constraints in solved() | 14:25 |
kanzure | I think I've lost you. | 14:25 |
fenn | as pure code, not some kind of meta-code | 14:26 |
fenn | if(condition1 && condition2 && (blah || blah2)) | 14:26 |
kanzure | I'm trying to come up with a way to express a 'range' | 14:26 |
fenn | not eval('if'('condition1' '&&' 'condition2' | 14:26 |
kanzure | so greater-than or less-than are important parts | 14:26 |
kanzure | right right | 14:26 |
kanzure | but those aren't really constraints | 14:26 |
kanzure | erm | 14:26 |
kanzure | I'm doing something a little different I think | 14:27 |
kanzure | you have an acceptable range for a physical quantity going in some direction | 14:27 |
kanzure | this 'acceptable range' is defined by some manipulation of the following operators: <, >, <=, >=, ==, +- | 14:27 |
kanzure | in some order possibly with &&, ||, etc. | 14:28 |
kanzure | do you know the typical newbie programming assignment? the one where you're asked to make a boolean algebra tree or something | 14:28 |
fenn | no | 14:28 |
kanzure | I don't recall how they go, but I'm fairly certain that's the same thing here. | 14:29 |
kanzure | hrm. | 14:29 |
kanzure | boolean expression tree | 14:29 |
fenn | it just does order of operations basically right? | 14:29 |
fenn | why reinvent the wheel | 14:29 |
fenn | any language already does that | 14:29 |
kanzure | so, isAcceptableRange() would be defined by each package maintainer? | 14:30 |
kanzure | for each interface definition | 14:30 |
fenn | no that's stupid | 14:31 |
kanzure | http://stackoverflow.com/questions/623413/expression-trees-for-dummies | 14:31 |
kanzure | then what are you suggesting | 14:31 |
fenn | build a list of requirements | 14:31 |
kanzure | (look for "Anyway, for an the expression" | 14:31 |
kanzure | ) | 14:31 |
kanzure | except with logical operators too | 14:31 |
kanzure | or, erm, only logical operators | 14:31 |
fenn | no, you need physical quantities and references to code variables too | 14:32 |
kanzure | physical quantities is taken care of in my model. | 14:32 |
kanzure | oh, but I guess there's no cross-over | 14:33 |
kanzure | interdependent variables I mean.. | 14:33 |
fenn | so there's two general ways to go about building this list, a bunch of functions, or a bunch of lambda's nested | 14:33 |
kanzure | crap. | 14:33 |
kanzure | I was assuming only independent variables | 14:33 |
fenn | note that functions/lambda's can contain any logic operator | 14:33 |
kanzure | what's a lambda? | 14:34 |
fenn | ... a function | 14:34 |
kanzure | ok | 14:34 |
fenn | mathy people like it for some reason | 14:34 |
fenn | probably because it's hard to read | 14:34 |
kanzure | so there are cases of interdependent variables? | 14:35 |
fenn | what does that mean | 14:35 |
kanzure | for instance, I was hoping to just say "range of N: 20 to 25 N" and then "range of kg: 5 to 9 kg" | 14:35 |
kanzure | and keep them separated | 14:35 |
kanzure | but it sounds like you want something more complicated? | 14:35 |
kanzure | so I was going to evaluate those in isolation of each other .. in other words, the range of one does not influence what possible range of the other could be | 14:36 |
fenn | if (N in range(20,25) and kg in range(5,9)): return true | 14:36 |
kanzure | right right | 14:36 |
kanzure | but it sounds like you want something more complicated than that | 14:36 |
kanzure | like "if N in range(20,25) but kg in range (6,7) out of the total range(5,9), then N has to be specifically something else" | 14:37 |
kanzure | or something. | 14:37 |
kanzure | erm. | 14:37 |
fenn | if (N in range(20,25) and kg in range(5,9) and kg = N*9.8) ? | 14:37 |
kanzure | I hope that's not the case | 14:37 |
kanzure | oh. um. | 14:37 |
kanzure | whatever. it sounds like I just misinterpreted you | 14:37 |
kanzure | if that's as complicated as you can imagine getting it | 14:37 |
kanzure | then that's good. | 14:37 |
fenn | no, i can imagine it much more complicated | 14:37 |
kanzure | uh oh. | 14:37 |
fenn | but i think python syntax is capable of handling it | 14:37 |
fenn | i just dont feel like typing out pages of example code right now | 14:38 |
kanzure | ok | 14:38 |
kanzure | btw, the reason why it can't be written out like that- as python code- is because it doesn't .. erm. | 14:38 |
kanzure | so, if you have two interfaces, both have definitions like that | 14:38 |
kanzure | and you want to check if the ranges are compatible between the two | 14:38 |
kanzure | just having two if()'s or two evaluation-range-functions-dealies | 14:39 |
kanzure | is not going to be useful | 14:39 |
fenn | if a.satisfied() and b.satisfied() isnt good enough? | 14:39 |
kanzure | how do you know it's satisfied | 14:40 |
fenn | because you wrote the damn function | 14:40 |
kanzure | you can't compare overlapping ranges | 14:40 |
kanzure | if two sets of bounds are within each other, or whatever, then good | 14:40 |
kanzure | but you can't check that if it's written out in code. | 14:40 |
kanzure | because then you'd need to run some sort of continuum checker solver | 14:40 |
fenn | yes you can | 14:41 |
kanzure | which is going to be nasty | 14:41 |
kanzure | ? | 14:41 |
fenn | a = buggy; b = baby | 14:41 |
kanzure | ? | 14:41 |
fenn | a.baby = b | 14:41 |
kanzure | ? | 14:41 |
-!- samrose [n=samrose@ip-207-145-38-45.iad.megapath.net] has quit [Remote closed the connection] | 14:41 | |
fenn | and in the buggy code you have some expression that checks whether the baby will fit | 14:41 |
fenn | s/buggy/basket/ | 14:41 |
kanzure | so it has to have its own checker? | 14:41 |
fenn | yes of course | 14:42 |
fenn | as part of the solver | 14:42 |
fenn | er, evaluator | 14:42 |
kanzure | who writes the code to check that a.baby fits? | 14:42 |
fenn | the buggy/basket maintainer | 14:42 |
fenn | or maybe it's just generic collision detection code stolen from elsewhere | 14:43 |
* fenn runs some errands | 14:46 | |
kanzure | errands.go() | 14:47 |
genehacker | got some errands to do too | 15:00 |
kanzure | argh the whole point of the internet is to get rid of geographic barriers | 15:04 |
kanzure | wtf is this.. | 15:04 |
* kanzure grins | 15:21 | |
kanzure | "Single cell transfection using plasmid decorated AFM probes" | 15:22 |
genehacker | haha | 15:23 |
kanzure | 30% success rate | 15:24 |
kanzure | well that's not good.. | 15:24 |
genehacker | 30% | 15:25 |
genehacker | ? | 15:25 |
genehacker | I 'd expect 100 | 15:25 |
genehacker | if they're putting it RIGHT ON THE CELL | 15:25 |
kanzure | "how can you possibly fuck that up" | 15:25 |
kanzure | "Bacterial cell curvature through mechanical controll of cell grwoth" | 15:30 |
genehacker | ??? | 15:31 |
kanzure | "Highly efficient molecular delivery into mammalian cells using CNT spearing". | 15:36 |
kanzure | heh. spearing. | 15:36 |
* kanzure feels like a caveman | 15:36 | |
genehacker | I loled | 15:37 |
genehacker | this one company sells spear tipped micromanipulators for catch and release of microorganisms | 15:38 |
kanzure | "Light-induced gene transfer from packaged DNA enveloped in a dendrimeric photosensitizer" | 15:40 |
kanzure | heh heh | 15:52 |
kanzure | neat. photoporation with violet LEDs | 15:52 |
kanzure | sub-milliwatt power and an exposure time of tens of milliseconds | 15:52 |
kanzure | 405 nm | 15:52 |
genehacker | hey I had a laser that was 405 nm | 15:53 |
kanzure | hey, do you know Ben-Yakar or Frederic Bourgeois? | 15:57 |
kanzure | they are working on this sort of thing apparently | 15:57 |
kanzure | here at UT | 15:57 |
genehacker | those names sound familiar | 15:57 |
kanzure | plus microfluidic integration | 15:58 |
genehacker | photoporation? | 15:58 |
genehacker | using light | 15:58 |
kanzure | well, no, they call it "laser nanosurgery" | 15:58 |
kanzure | but that's the same damn thing | 15:58 |
genehacker | what cells did they transfer genes to? | 15:58 |
* kanzure reads | 15:58 | |
kanzure | C. elegans | 15:59 |
kanzure | I don't think they actually did it, they were just thinking about it | 15:59 |
kanzure | oh, nevermind | 15:59 |
kanzure | they did. | 15:59 |
genehacker | oh that paper5 | 16:00 |
kanzure | serial trapping of worms in channels, etc. | 16:00 |
kanzure | hrm. so immobilize cells in a gel. transfect with LED photoporation. not bad. | 16:02 |
genehacker | animal cells | 16:07 |
genehacker | photoporation | 16:07 |
genehacker | hmmm | 16:07 |
kanzure | what was the name of that paper that used LEDs to produce millions of degrees of heat in a bubble? | 16:09 |
kanzure | ah, "Whole Blood Pumped by Laser Driven Micropump" | 16:10 |
kanzure | oh. makes sense. I was looking that up for "cell lysis". | 16:12 |
kanzure | so of course it should work for transfection.. | 16:12 |
genehacker | kanzure do me a favor and find me a micropump that is 90% efficient | 16:13 |
kanzure | hm. weird. DNA synthesis with only three LEDs and a capillary tube? | 16:33 |
kanzure | http://heybryan.org/books/papers/A%20scalable%20method%20for%20multiplex%20LED-controlled%20synthesis%20of%20DNA%20in%20capillaries.pdf | 16:34 |
kanzure | ack. | 16:36 |
kanzure | 180 s for photodeprotection and 60 sec for coupling of the base | 16:36 |
kanzure | 360 seconds per base | 16:36 |
kanzure | 160 nt in 16 hours. | 16:37 |
kanzure | I guess that's not too terrible :-/ | 16:37 |
-!- nsh [n=nsh@wikipedia/nsh] has joined #hplusroadmap | 16:37 | |
-!- cis-action [n=cis-acti@c-67-180-9-222.hsd1.ca.comcast.net] has joined #hplusroadmap | 17:06 | |
-!- cis-action [n=cis-acti@c-67-180-9-222.hsd1.ca.comcast.net] has quit [Client Quit] | 17:06 | |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has joined #hplusroadmap | 17:22 | |
kanzure | hey cis-action. | 17:26 |
kanzure | I just sent out a list of really neat papers to diybio | 17:26 |
kanzure | including transfection of cells via LEDs | 17:26 |
-!- nsh- [n=nsh@host86-133-35-168.range86-133.btcentralplus.com] has joined #hplusroadmap | 17:46 | |
fenn | the LED thing is totally obvious but still really cool | 18:01 |
kanzure | neat image: http://heybryan.org/books/papers/Nanoscale%20Operation%20of%20a%20Living%20Cell%20Using%20an%20Atomic%20Force%20Microscope%20with%20a%20Nanoneedle-fig4.png | 18:02 |
fenn | you could make your own PCR primers | 18:02 |
kanzure | hm? | 18:02 |
kanzure | oh, with the synthesizer | 18:02 |
kanzure | yes | 18:02 |
fenn | poor cells | 18:03 |
kanzure | RAPE | 18:03 |
fenn | heh that'd be fun to heckle people with at a biotech conference | 18:03 |
-!- nsh [n=nsh@wikipedia/nsh] has quit [Connection timed out] | 18:03 | |
kanzure | fenn: now we just need to get a fundie to believe it. | 18:04 |
fenn | huh. you know, those capillary tubes are the same as i used in the PCR machine at IU | 18:05 |
kanzure | instead of me spitting out tons of links, here's some latest papers: http://heybryan.org/books/papers/?C=M;O=D | 18:05 |
fenn | so you could feasibly put the PCR mix, nucleotides, synthesis, and amplification sequence all in the same tube/machine | 18:05 |
kanzure | hum. | 18:06 |
kanzure | sounds like it. | 18:06 |
-!- any49654908 [n=someone@75-120-24-167.dyn.centurytel.net] has joined #hplusroadmap | 18:06 | |
fenn | well, not really | 18:06 |
kanzure | it would be staged operation | 18:06 |
fenn | they have to wash the tube between steps right? | 18:06 |
kanzure | have to inject the strand you want to amplify later | 18:06 |
kanzure | um | 18:06 |
kanzure | I didn't see anything about that | 18:06 |
kanzure | I'm still kind of confused by it | 18:06 |
kanzure | why do they have multiple LEDs? is there diffusion going on here? | 18:06 |
fenn | no, the synthesized oligo is bound to the capillary wall | 18:07 |
kanzure | how would the light reach the strands being built way the hell on the other side of the tube | 18:07 |
kanzure | oh | 18:07 |
fenn | th idea is you have multiple synthesis sites along the tube (one per LED) | 18:07 |
kanzure | right | 18:07 |
fenn | with as sharp a cutoff as possible to reduce half-formed strands | 18:07 |
-!- katsmeow-afk [n=someone@75-121-60-222.dyn.centurytel.net] has quit [Read error: 60 (Operation timed out)] | 18:08 | |
kanzure | so you wash it in between steps or not? | 18:08 |
kanzure | how do you change the nucleotide out? and what's the point of having 3 LEDs? why not just one | 18:08 |
fenn | dunno | 18:09 |
kanzure | it's neat to find these papers hidden in the literature | 18:10 |
fenn | since you can't have a perfect sphere (no way to wash it out) i guess you're stuck with half-formed product anyway | 18:10 |
kanzure | nobody noticed them before as being all too important because, honestly, the performance /does/ suck | 18:10 |
kanzure | but it's better than nothing | 18:10 |
fenn | yeah certainly, good enough for pcr primers | 18:10 |
fenn | once you have a bottle of the right chemicals you get essentially as many pcr primers as you want | 18:10 |
fenn | which is way better than $20/pop | 18:11 |
kanzure | did you see the LED-based photoporation paper? | 18:12 |
kanzure | or the plasmid-covered AFM probe transfection method? | 18:13 |
fenn | 'this method is comparable or surpasses the quality of published commercial oligonucleotides' | 18:23 |
fenn | so i suggested meeting somewhere at 4, the guy replied 'ok' so i assumed he'd show up | 18:26 |
fenn | but then the next day at 11 am he wanted a confirmation, but i was asleep, so he never showed up! | 18:26 |
kanzure | hm. | 18:26 |
fenn | is it standard protocol to respond 'ok' to 'ok'? | 18:27 |
kanzure | not to my knowledge | 18:27 |
kanzure | another neat image: http://heybryan.org/books/papers/Direct%20observation%20of%20DNA%20translocation%20and%20cleavage%20by%20the%20EcoKI%20endonuclease%20using%20AFM.pdf.1.png | 18:31 |
kanzure | an endonuclease in action | 18:31 |
* fenn mumbles something about .jpg's | 18:38 | |
kanzure | wha? | 18:38 |
kanzure | they were png files | 18:38 |
kanzure | I don't like jpegs because then I have to decide on compression | 18:38 |
fenn | oh boo hoo | 18:39 |
fenn | so no compression at all is better? | 18:39 |
kanzure | I'm an idiot, aren't I? | 18:39 |
kanzure | :/ | 18:39 |
genehacker | yeah | 18:39 |
genehacker | JPEG sux | 18:39 |
fenn | it's better than PNG for photographs | 18:39 |
genehacker | especially for things with big borders | 18:39 |
genehacker | like webcomics | 18:40 |
fenn | borders? | 18:40 |
fenn | like a demotivational poster? | 18:40 |
genehacker | ie MSpaint type drawings | 18:40 |
genehacker | exactly | 18:40 |
genehacker | http://images.google.com/images?q=compression&oe=utf-8&rls=org.mozilla:en-US:official&client=firefox-a&um=1&ie=UTF-8&sa=N&hl=en&tab=wi | 18:41 |
genehacker | see the compression poster | 18:41 |
fenn | eh, what? | 18:41 |
genehacker | http://emmastott.me/blog/tag/comrpession/ | 18:41 |
genehacker | to get the full effect, you have to see the poster downsized first | 18:41 |
fenn | whatever.. those should be .svg anyway | 18:42 |
kanzure | another neat image: RNA polymerase doing its thing: http://heybryan.org/books/papers/Direct%20observation%20of%20one-dimensional%20diffusion%20and%20transcription%20by%20Escherichia%20coli%20RNA%20polymerase.pdf.1.jpeg | 18:43 |
fenn | how do they get these pictures? | 18:44 |
kanzure | AFM | 18:45 |
kanzure | hrm.. single-plasmid PCR with an AFM tip made of mica. hrm. and a mildly acidic solution. silicon nitride tip. | 18:45 |
genehacker | mica? | 18:46 |
genehacker | I think I found some of that once | 18:46 |
genehacker | interesting stuff | 18:46 |
kanzure | single plasmid PCR with AFM tip: http://heybryan.org/books/papers/Recovery%20and%20amplification%20of%20plasmid%20DNA%20with%20AFM%20and%20PCR%20-%20single-plasmid%20PCR.pdf | 18:49 |
fenn | http://www.youtube.com/watch?v=V1qWmrp6GAs Carbon atoms moving at the edge of a hole in graphene0:16 | 18:49 |
fenn | Carbon atoms moving at the edge of a hole in graphene been directly observed in real time. With a transmission electron | 18:49 |
kanzure | heh. so couple that with the AFM-based plasmid delivery system. | 18:49 |
fenn | blarg | 18:49 |
kanzure | you fail at copy-paste | 18:49 |
* fenn blames youtube | 18:49 | |
fenn | i thought afm needed vacuum, not warm sticky cell gloop | 18:50 |
kanzure | so, if you can use an AFM tip to deliver plasmids into a cell, and amplify a single plasmid on a tip, .. | 18:50 |
kanzure | apparently not? | 18:50 |
kanzure | I've been seeing some room-temperature stuff | 18:51 |
kanzure | it would be fun to grow a collection of microbes and randomly pit them against each other and take AFM imaging videos of the results | 18:52 |
kanzure | since there's such a ridiculously large diversity of them, you'll never grow bored | 18:52 |
fenn | except 99% of them will just ignore each other | 18:52 |
kanzure | then poke a hole in one of them and place the other in it | 18:53 |
kanzure | the survivor, wins! | 18:53 |
kanzure | heh' single-cell directed evolution | 18:53 |
kanzure | uh oh, transcellularism? | 18:53 |
fenn | in high school we had an 'ant gladiator arena' | 18:53 |
fenn | my ants always lost | 18:54 |
kanzure | :( | 18:54 |
fenn | i went to fry's electronics today, way the hell out there | 18:55 |
kanzure | yep. | 18:55 |
fenn | they actually have real electronics tools and components and stuff | 18:55 |
kanzure | the home I was going to be put into is right around the corner | 18:55 |
kanzure | convenient, but not. | 18:55 |
fenn | not | 18:56 |
kanzure | so I'm not sure now whether to keep reading maniacally or not, | 18:57 |
kanzure | photoporation + dna synthesis + afm stuff kind of completes the chain | 18:57 |
fenn | why is photoporation better than other stuff? dont you need (yet more expensive) chemicals? | 19:03 |
kanzure | huh? it looked like you just shine a focused LED at a cell | 19:04 |
kanzure | and the plasmids go in. | 19:04 |
kanzure | kind of like the cell lysis method with LEDs except less intense | 19:04 |
fenn | pew pew | 19:05 |
kanzure | transfection by impingement? | 19:05 |
kanzure | or is that the sound of a laser? | 19:05 |
fenn | don't you need a microscope? | 19:05 |
kanzure | not if you know where the cells are | 19:06 |
kanzure | or, erm, maybe for the lense | 19:06 |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has quit [] | 19:07 | |
kanzure | (I sent the link in the recent diybio email) | 19:07 |
kanzure | ah: http://heybryan.org/books/papers/Violet%20diode-assisted%20photoporation%20and%20transfection%20of%20cells.pdf | 19:07 |
fenn | der.. 'focused at a spot 1mm in diameter' | 19:08 |
fenn | is that right? | 19:08 |
fenn | no, it's 1um | 19:08 |
fenn | wtf is she using a camera shutter | 19:09 |
kanzure | yes | 19:09 |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has joined #hplusroadmap | 19:10 | |
fenn | have you found any references of photoporation with bacteria? | 19:11 |
kanzure | haven't looked | 19:12 |
kanzure | doesn't look like there's anything. | 19:14 |
fenn | i bet they are too small for it to work | 19:14 |
fenn | did ultrasound work with bacteria? | 19:15 |
fenn | yes, looks like it | 19:15 |
kanzure | ultrasound work with bacteria for what? | 19:15 |
kanzure | also, there was another paper- "Sonoporation of suspended cells with a single cavitation bubble in microfluidic confinement" | 19:16 |
kanzure | http://heybryan.org/books/papers/Sonoporation%20of%20suspension%20cells%20with%20a%20single%20cavitation%20bubble%20in%20a%20microfludici%20confinement.pdf | 19:16 |
kanzure | but that was for hman HL60 cells | 19:16 |
kanzure | "cavitation bubbles can induce membrane poration of cells located in their close vicinity" | 19:17 |
fenn | The mixture was subjected to 1-MHz US treatment under 0.5 MPa for 90 seconds, with a duty cycle of 50% | 19:17 |
fenn | that sounds doable | 19:17 |
fenn | hard to imagine it being any cheaper than electroporation though | 19:18 |
fenn | looks like it's much more efficient and higher cell survival rate | 19:18 |
fenn | not like that really matters with bacteria | 19:19 |
kanzure | "We demonstrate the ability to accurately position up to 34 micrometer sized bubbles using laser energies of 56 microjoules" using a spatial light modulator (SLM) (a.k.a, LCD hacks here we come) <- generation of laser-induced cavitation bubbles with a digital hologram (so, not ultrasound) | 19:19 |
kanzure | this is getting a bit too complicated for me to remember all at once | 19:19 |
kanzure | neat, they write stuff out of bubbles | 19:20 |
fenn | interesting "Sonoporation may activate the stress response genes, including RecA, thus increasing the frequency of double-crossover homologous recombination." | 19:20 |
kanzure | Generation of laser-induced cavitation bubbles with a digital hologram http://heybryan.org/books/papers/Generation%20of%20laser-induced%20cavitation%20bubbles%20with%20a%20digital%20hologram%20-%20LCD%20-%20spatial%20light%20modulator.pdf | 19:21 |
fenn | i'm surprised you can't buy a real holographic display yet | 19:22 |
fenn | with some of this GPGPU stuff it should be easy | 19:22 |
kanzure | ever hear of sonoluminescence? | 19:22 |
fenn | yes | 19:22 |
fenn | only at high power levels | 19:22 |
genehacker | sonolumination is weird | 19:24 |
genehacker | some people think it might be fusion | 19:24 |
fenn | genehacker: other way around | 19:25 |
fenn | fusion might be caused by sonoluminescence | 19:25 |
genehacker | yeah | 19:26 |
fenn | wow 1024x768 LCD in a 20mm package | 19:27 |
genehacker | WHERE! | 19:29 |
genehacker | OH GOD I CAN SEE FOREVER | 19:30 |
genehacker | I am not talking about the LCD | 19:30 |
fenn | are you all right? | 19:30 |
genehacker | not really | 19:34 |
genehacker | mental stability is ok | 19:34 |
genehacker | it's just slightly disturbing | 19:34 |
genehacker | a movie called moonwalker | 19:34 |
genehacker | exists | 19:35 |
genehacker | let's see that LCD screen now shall we? | 19:35 |
fenn | http://www.holoeye.com/lcos_microdisplay_color_sequential.html | 19:36 |
genehacker | oh that one | 19:36 |
fenn | it's very similar to the spatial light modulator they're using in that last paper | 19:37 |
genehacker | we need one that is black and white | 19:37 |
genehacker | and that we can put light through | 19:37 |
genehacker | we can't use an LCD in it's reflective mode | 19:37 |
fenn | http://www.holoeye.com/spatial_light_modulator_lc_r_2500.html | 19:37 |
fenn | it's not reflective, it's transmissive | 19:37 |
fenn | er. i think it is at least | 19:38 |
genehacker | not the LCD display | 19:38 |
genehacker | or at least one of them is | 19:38 |
genehacker | that SLM is cool | 19:38 |
fenn | ok i'm wrong, they're both reflective | 19:38 |
fenn | in which case, why is it special for sequential color | 19:39 |
fenn | *drool* | 19:39 |
kanzure | so uhh | 19:46 |
kanzure | if you can do single plasmid AFM tip stuff, | 19:46 |
kanzure | why not single DNA strand AFM tip stuff | 19:47 |
kanzure | and then why not ligate strands of DNA together by pick and place? | 19:47 |
fenn | to do what? | 19:48 |
kanzure | very large genome synthesis | 19:49 |
fenn | why is that better than sticky end ligation | 19:50 |
fenn | (self assembly) | 19:50 |
kanzure | you could have many, many strands of DNA on a surface | 19:51 |
kanzure | and then you assemble them in order by picking them up | 19:51 |
kanzure | and then doing some facy ligation only once at the assembly site for attaching each strand | 19:51 |
kanzure | (maybe some of that fancy ultrasound-picoliter-stuff) | 19:51 |
fenn | picoliter is way huge compared to a dna strand | 19:51 |
kanzure | eh? | 19:52 |
kanzure | crap crap crap. | 19:52 |
fenn | you could probably fit a human genome in a picoliter | 19:52 |
fenn | cube 10 microns on a side | 19:53 |
genehacker | kanzure, just be cause we can doesn't mean we should | 19:53 |
genehacker | just because we can doesn't mean it's efficient | 19:53 |
kanzure | speak for yourself mr. fluidic replicator | 19:54 |
kanzure | just trying to find a way to avoid chemicals. | 19:55 |
kanzure | which is awkward since it's all chemistry | 19:55 |
genehacker | i don't have to, exponential growth | 19:55 |
fenn | just jamming dna together with a stick isn't going to covalently bond it | 19:56 |
fenn | at least not the way you want it to | 19:56 |
fenn | so you're going to need ligase anyway | 19:56 |
kanzure | right | 19:56 |
kanzure | I'm ok with having to use ligase | 19:56 |
kanzure | can you ligate two nucleotides together? <- there you go.. | 19:57 |
fenn | you could synthesize both strands of a short sequence (100bp or whatever) but leave overlapping sticky ends | 19:57 |
kanzure | right. traditionally you just attach that with a ligase step to the bigger piece that you're currently synthesizing | 19:58 |
fenn | then when you anneal, the matching oligos will pair up first, then the matching sticky ends will pair up | 19:58 |
fenn | oh, i'd do it all at once | 19:58 |
kanzure | I thought sticky ends would .. | 19:58 |
kanzure | erm. how are sticky ends addressed together? | 19:58 |
fenn | complementary base pairing | 19:59 |
genehacker | mechanosynthesis | 19:59 |
kanzure | there's only so many base pairs, so you can only ligate two strands together at once | 19:59 |
kanzure | two different strands I mean | 19:59 |
genehacker | you're doing it wrong | 19:59 |
fenn | hang on lemme draw a pic | 19:59 |
fenn | genehacker: shut up | 19:59 |
genehacker | ok | 19:59 |
fenn | thanks :) | 19:59 |
fenn | http://imagebin.org/46044 | 20:01 |
fenn | each sticky sequence would be different | 20:02 |
fenn | there would be a maximum number of oligos you could anneal at the same time before you start getting mismatches | 20:02 |
kanzure | what is blue, what is red? | 20:03 |
kanzure | complementary strands? | 20:03 |
fenn | you pick the ligation sites for maximum selectivity | 20:03 |
fenn | yes | 20:03 |
fenn | synthesized in different spots on the light directed synthesis | 20:03 |
kanzure | I still don't see why annealed oligos would necessarily form in some order for the final strand.. | 20:04 |
fenn | it's a grammar | 20:04 |
fenn | a sticks to ~a, b sticks to ~b | 20:05 |
kanzure | so then how do you get a to stick to b? | 20:05 |
fenn | you don't | 20:05 |
kanzure | then how do you get an assembled molecule | 20:05 |
fenn | you put a and ~b on one strand, b and ~a on the other ( that would form a ring) | 20:05 |
fenn | a and b are the sticky sequence | 20:06 |
fenn | you leave that part out of the complementary strand so it's exposed as single strand DNA | 20:06 |
kanzure | so you're using complementary strands of ssDNA tails or something? | 20:07 |
kanzure | and these tails (ideally) match up more often than errors? to be ligated together | 20:07 |
fenn | yes, this is basic recombinant DNA stuff | 20:07 |
kanzure | blah, didn't know it was watson-crick base pairing crap | 20:07 |
fenn | in old fashioned techniques they use restriction enzymes to cut a genome or plasmid at the sticky sequence | 20:08 |
fenn | but restriction enzymes only recognize like 5-15bp | 20:08 |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has quit [] | 20:08 | |
fenn | whereas i think you'd want a longer sequence so you could have more oligo's in the same annealing step | 20:08 |
fenn | so at the end you will have a sequence of a certain length, run it on a gel to remove all the mismatched sequences (should be able to discriminate by the number of bp in your oligo) | 20:10 |
kanzure | Mechanical separation of the complementary strands of DNA. "the typical forces along the opening are in the range of 10-15 pN. The separation force signal is shown to be related to the local GC vs. AT content along the molecule." | 20:10 |
kanzure | "variations of this content on a typical scale of 100-500 bases are presently detected." (1997) | 20:10 |
fenn | did you ever compile pythonocc "--with-doc"? | 20:15 |
kanzure | no | 20:15 |
kanzure | have you? | 20:15 |
fenn | i'm wondering if it's simple (and why isnt it on by default??) | 20:15 |
kanzure | it might be the doxygen stuff | 20:15 |
fenn | no | 20:16 |
fenn | it adds docstrings | 20:16 |
kanzure | gasp | 20:16 |
kanzure | docstrings? is that the stuff before a function that describes it? | 20:16 |
fenn | then you can just do help(foo) or pydoc foo | 20:16 |
fenn | yes | 20:16 |
kanzure | because I get popups in InteractiveVeiwer that tells me about the functions, in terms of parameters | 20:16 |
kanzure | oh | 20:16 |
kanzure | but not full text | 20:16 |
kanzure | okay, guess it's different | 20:16 |
fenn | it's supposed to be helpful natural language text | 20:16 |
fenn | somehow i doubt OCC is going to be helpful though :P | 20:17 |
drazak | kanzure: | 20:23 |
drazak | I haven't read any of the openthermocycler thread | 20:23 |
drazak | but | 20:23 |
drazak | have peltier devices been suggested? | 20:23 |
drazak | along with adding drops of mineral oil to the microcuvettes? | 20:24 |
-!- duzt [i=duzt__@63.sub-75-204-238.myvzw.com] has joined #hplusroadmap | 20:24 | |
kanzure | drazak: not anything about mineral oil, IIRC | 20:27 |
drazak | it will prevent condensation from building on the top of the device from uneven heating, you have to make sure to pipette from underneath the layer of oil when removing samples | 20:28 |
drazak | peltier devices are simple DC heaters/coolers that could be efficiently used | 20:28 |
-!- DrTread [n=irchon@166.133.152.12] has joined #hplusroadmap | 20:29 | |
kanzure | Hey DrTread. | 20:30 |
kanzure | DrTread: Creative uses of LEDs for diybio: http://groups.google.com/group/diybio/t/ba6de9183f9ba83e | 20:30 |
drazak | led's are good for RTPCR too | 20:31 |
DrTread | hello. I'm @ friend's house, bored | 20:31 |
DrTread | saw that on your twitted. :-) | 20:31 |
kanzure | drazak: feel free to reply to that thread on diybio | 20:31 |
kanzure | DrTread: woah, you actually read my twitter feed? :) | 20:32 |
drazak | kanzure: I will soon | 20:32 |
kanzure | drazak: I mean the one about LEDs :) | 20:32 |
DrTread | I hang on every word! | 20:32 |
drazak | I'll post them both | 20:33 |
drazak | I'm thinking about other things too :) | 20:36 |
drazak | the people who are doing the open thermocycler should include a buffer | 20:36 |
drazak | that you cna buy more of | 20:36 |
DrTread | kanzure, you mentioned the collyer brothers. I'm watching a movie about other famous recluses | 20:36 |
kanzure | DrTread: Is it a good movie? | 20:36 |
DrTread | the Edies Bouvier | 20:37 |
DrTread | documentary. weird. | 20:37 |
DrTread | see link to them @ end of collyer's wikipedia | 20:37 |
DrTread | tonight there's a movie on HBO about them. | 20:38 |
DrTread | these people male me want to give away all of my junk | 20:39 |
kanzure | turn it off. | 20:39 |
kanzure | before damage occurs | 20:39 |
DrTread | Pffft. I'm at a friend's house. I can't turn of off. :-( | 20:40 |
DrTread | I don't watch any TV. I don't know what to do. :( | 20:40 |
genehacker | DrTread just don't watch moonwalker and you'll be alright | 20:42 |
DrTread | no problem. ;) | 20:42 |
drazak | kanzure: If I send an email to the email I get digests from, will it post to the board? | 20:44 |
fenn | genehacker: there was a 'moonwalker' video game, where you were michael jackson, and the goal was to collect blonde children :) | 20:44 |
fenn | and avoid ninjas with machine guns | 20:45 |
genehacker | oh god | 20:45 |
fenn | there was even a sequel | 20:45 |
genehacker | I just know that in the movie michael jackson morphs into a jet, then a flying car, then a jack rabbit | 20:45 |
genehacker | yes there were children | 20:45 |
fenn | drazak: you can avoid having to use mineral oil with a heated lid | 20:46 |
DrTread | MJ is unfettered by good taste. | 20:46 |
drazak | fenn: right | 20:46 |
drazak | fenn: but if we're making this the cheapest thermocycler out there, why have a heated lid? | 20:47 |
drazak | not that a second peltier device/water chamber will cost much, but... | 20:47 |
fenn | because a heated lid is so ridiculously easy it'd be stupid not to | 20:47 |
fenn | you dont even need a peltier, just some power resistors | 20:47 |
DrTread | yes, heat is easy. | 20:48 |
fenn | i'm really wondering why you need a peltier at all | 20:48 |
drazak | peltier is 1000x more efficient, you can cool with it, and it has more surface area | 20:48 |
fenn | wouldnt a simple heatsink+fan work better? | 20:48 |
drazak | you can cool with a peltier | 20:48 |
drazak | flip the current | 20:48 |
fenn | but PCR is 50-95C most of the time | 20:48 |
drazak | right | 20:48 |
fenn | who gives a flying fuck about cooling | 20:48 |
drazak | but you have to go from 95C to 50C at one step | 20:48 |
-!- DrTread [n=irchon@166.133.152.12] has quit [Remote closed the connection] | 20:49 | |
drazak | if you're doing 20 or 30 cycles | 20:49 |
fenn | and that's when you turn the fan on | 20:49 |
drazak | and it takes 5 minutes to cool by hair | 20:49 |
fenn | bah | 20:49 |
drazak | that's an extra 90 minutes or so | 20:49 |
-!- DrTread [n=irchon@166.133.152.12] has joined #hplusroadmap | 20:49 | |
fenn | you have to get rid of that heat somehow | 20:49 |
drazak | and then the time spent /at/ the temperatures | 20:49 |
drazak | which is why you have an upper and lower water bath | 20:50 |
fenn | water bath??? | 20:50 |
drazak | yup | 20:50 |
fenn | GTFO | 20:50 |
drazak | to keep consistant temperature over each microcuvette | 20:50 |
drazak | submerse the bottoms of each in a water bath | 20:50 |
drazak | hey, this is my idea | 20:50 |
drazak | I haven't read the thread :P | 20:50 |
DrTread | big blocks of aluminum work great. | 20:50 |
drazak | DrTread: sure, if you want some cuvettes to work and some not to | 20:51 |
drazak | which means more reagent | 20:51 |
drazak | which is the expensive part, in the end | 20:51 |
fenn | drazak: almost all commercial PCR machines use big blocks of aluminum | 20:51 |
drazak | huh | 20:51 |
drazak | it's prone to uneven heating, imo | 20:51 |
fenn | not water baths | 20:51 |
fenn | you're wrong, sorry | 20:51 |
drazak | well what do I know? :P | 20:51 |
drazak | but they are prone to uneven heating, aluminium blocks | 20:51 |
drazak | you don't have to be an ass about my being wrong | 20:52 |
fenn | aluminum is the third most conductive element | 20:52 |
fenn | sorry if i'm being an ass | 20:52 |
drazak | well then | 20:52 |
fenn | but water bath is just so wrong.. | 20:52 |
drazak | your heater has to have enough surface area to evenly heat it | 20:52 |
drazak | also | 20:52 |
drazak | power resistors are slow, no? | 20:53 |
DrTread | single xtal diamond is great, but expensive | 20:53 |
fenn | sorry, gold is third, aluminum is fourth | 20:53 |
fenn | by area | 20:53 |
DrTread | best way is to pump temp controlled water thru alum block. | 20:53 |
drazak | mhm | 20:54 |
DrTread | of | 20:55 |
DrTread | course, conduction to cuvettes is an issue | 20:55 |
fenn | hey i know let's submerse them in a pool of mercury | 20:56 |
* drazak eyerolls | 20:56 | |
DrTread | LOL! no, Na-K | 20:57 |
fenn | you could dispense with the heating element altogether and resistively heat the molten metal | 20:57 |
* drazak eyerolls | 20:57 | |
fenn | ok ok how about field's metal | 20:57 |
fenn | it's "non-toxic" | 20:57 |
drazak | galium? | 20:57 |
fenn | (note the quotes) | 20:58 |
drazak | we could use gallium resistively heated | 20:58 |
drazak | :D | 20:58 |
DrTread | yeah. that, too. buy it sticks to everything. | 20:58 |
fenn | i'm from the light bulb + capillary tube school of PCR | 20:59 |
drazak | that's slow though | 20:59 |
fenn | so worrying about heat conduction to plastic tubes seems silly when you can get a 10-fold improvement by switching to capillary tubes | 20:59 |
fenn | drazak: cycle time under a minute, usually | 20:59 |
drazak | :o | 20:59 |
drazak | how do the capillary tubes work then? | 20:59 |
drazak | also brb | 20:59 |
fenn | there's a glass tube, about a mm.. you stick it into your eppendorf (for large arrays of reactions i just dotted things out on parafilm) and it sucks the mix up into the tube.. then you seal the ends with a bunsen burner and stick it in a hole poked through a piece of rubber | 21:01 |
fenn | tube = 1mm OD | 21:01 |
fenn | inside the machine there's a light bulb, a fan, and a temperature sensor with the same thermal characteristics as the glass tubes | 21:01 |
fenn | all the tubes are the same distance from the bulb | 21:01 |
kanzure | hrm. comB2 and comB3 genes look interesting. | 21:01 |
drazak | fenn: seems tedious to setup | 21:02 |
fenn | drazak: this is the idaho scientific thermocycler, btw | 21:02 |
kanzure | five proteins in Heliobacter pylori form a transmembrane channel that induces natural competence. | 21:03 |
kanzure | hellooo | 21:03 |
fenn | http://www.idahotech.com/RapidCycler2/index.html | 21:03 |
DrTread | well, | 21:04 |
DrTread | let's give ourselves ulcers | 21:04 |
kanzure | doesn't matter, we just want the genes | 21:05 |
kanzure | does anyone here have ulcers and not mind gi surgery by an amateur? | 21:05 |
DrTread | you??? | 21:06 |
kanzure | in particular: VirB7, VirB8, VirB9, VirB10 | 21:06 |
fenn | i suggest not culturing human pathogens | 21:06 |
kanzure | oh ok | 21:07 |
fenn | i'm sure there are plenty of other naturally competent cells | 21:07 |
DrTread | heliobacter is hard to culture | 21:08 |
DrTread | anthrax is easy :-D | 21:08 |
kanzure | yay, these genes are sequenced and in NCBI | 21:08 |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has joined #hplusroadmap | 21:09 | |
kanzure | cis-action: hey | 21:09 |
kanzure | cis-action: you remember ADP1? | 21:09 |
cis-action | hi yep | 21:09 |
cis-action | just talking about it actually | 21:09 |
cis-action | ....why? | 21:09 |
kanzure | cis-action: natural competence is caused by a transmembrane channel protein complex (thingy) which happens to be sequenced and in NCBI | 21:09 |
kanzure | which potentially means us transplanting it to other organisms | 21:10 |
drazak | that thingy sounds like a connexin | 21:10 |
drazak | :D | 21:10 |
cis-action | ... suggesting we should just biobrick that up | 21:10 |
kanzure | hell yeah | 21:10 |
cis-action | how long is the gene(s) | 21:10 |
fenn | kanzure: "five proteins" meaning five different proteins or a 5-mer? | 21:10 |
kanzure | haven't counted :p but I'm looking at one that has a ~30 AA sequence | 21:10 |
kanzure | fenn: I think it's five different proteins, yes | 21:10 |
fenn | oh. that sucks | 21:10 |
kanzure | why? | 21:11 |
fenn | i bet there's something with just one gene product | 21:11 |
kanzure | they are on agrobacterium tumefaciens plasmid pTil5955 | 21:11 |
genehacker | ahem | 21:11 |
genehacker | I think we are forgetting the lightbulb thermocycler | 21:11 |
drazak | eh oh well | 21:11 |
fenn | nopaline? | 21:11 |
kanzure | VirB7, VirB8, VirB9, and VirB10. eh, which is only four. wait a moment. | 21:11 |
genehacker | aluminum isn't that expensive | 21:11 |
kanzure | okay. the paper says four. | 21:12 |
fenn | genehacker: it's not the expense, it's about the mess and inconvenience of a water bath | 21:12 |
genehacker | aluminum is hard to drill | 21:12 |
fenn | oh come on | 21:12 |
genehacker | if you won't have a drill press | 21:12 |
genehacker | like me | 21:12 |
fenn | i will drill some holes in a block for you | 21:13 |
DrTread | I have all tools | 21:13 |
drazak | that sounds so naughty | 21:13 |
genehacker | you have a drill press? | 21:13 |
DrTread | dirty minds think alike | 21:13 |
fenn | there are like 7 drill presses in the shop | 21:13 |
kanzure | this would make quite the hell of an igem project. hrm.. | 21:13 |
genehacker | then drill me some extruder barrels | 21:13 |
drazak | you know | 21:14 |
DrTread | I have a lathe, mill, too | 21:14 |
drazak | drill press seems like the wrong way to get holes in the aluminium for a cuvette | 21:14 |
drazak | you'd want something that pretty much exactly fit it | 21:14 |
fenn | drazak: yes, a reamer is probably better | 21:14 |
drazak | yeah | 21:14 |
fenn | i'd like to do some experimenting to see if it really matters | 21:14 |
genehacker | a liquid galium bath would be interesting | 21:15 |
genehacker | i have some liquid gallium | 21:15 |
fenn | step-drilling can get pretty good fits (makes a pop noise when you pull it out) | 21:15 |
DrTread | remember differential expansion | 21:15 |
fenn | DrTread: which expands more, LDPE or aluminum? | 21:15 |
genehacker | ah yes thermal expansion | 21:16 |
genehacker | aluminum | 21:16 |
fenn | or whatever eppendorf tubes are made of | 21:16 |
genehacker | aluminum | 21:16 |
drazak | make it a micron bigger than! | 21:16 |
DrTread | oh, I thought glass. n/m | 21:16 |
genehacker | well let me check thermal expansion coefficients | 21:16 |
drazak | I can't imagine going from ~298K-373K is a lot of expansion for Al | 21:16 |
fenn | microcentrifuge tubes are made of polypropylene | 21:17 |
fenn | or polycarbonate | 21:17 |
DrTread | ldpe is much bigger | 21:17 |
genehacker | how big are EP tubes drazak? | 21:17 |
drazak | not big | 21:18 |
DrTread | 7-8 mm od, iirc | 21:18 |
drazak | yeah | 21:18 |
drazak | but they come to a tip | 21:18 |
fenn | there are smaller, thin-wall tubes for PCR | 21:18 |
genehacker | http://en.wikipedia.org/wiki/Coefficient_of_thermal_expansion#Thermal_expansion_coefficients_for_some_common_materials | 21:18 |
genehacker | all be | 21:18 |
fenn | thanks lazyweb | 21:18 |
genehacker | look at the thermal expansion coefficient of PVC and look at aluminum | 21:19 |
drazak | yeah, those are the ones I was thinking of | 21:19 |
drazak | I think they're called microcuvettes | 21:19 |
drazak | they usually come in sheets, in real lab ones | 21:19 |
genehacker | yeah I know what you're talking about | 21:19 |
DrTread | ok, the main feature us starting. I'll sign off. | 21:20 |
genehacker | so what are EP tubes made of? | 21:21 |
fenn | PE+PP have huge thermal expansion, which means they'll swell up and get stuck in the holes at high temp (which is a good thing) | 21:21 |
drazak | PE? | 21:21 |
fenn | genehacker: polypropylene | 21:21 |
fenn | for the 'living hinge' | 21:21 |
genehacker | ok | 21:21 |
-!- DrTread [n=irchon@166.133.152.12] has quit [Remote closed the connection] | 21:22 | |
genehacker | so we want them to stay in the hole? | 21:22 |
fenn | when the tube expands, that pushes it against the walls | 21:22 |
fenn | which makes for good thermal contact | 21:22 |
fenn | also the lid will be pushing it down, so if the hole has a matching conical bottom it will push along the taper too | 21:23 |
fenn | i dont think most diy people will be up for making conical drill bits though | 21:23 |
genehacker | yeah me too | 21:23 |
fenn | (even though it's not really that hard) | 21:23 |
genehacker | we could just use gallium | 21:26 |
genehacker | though gallium likes to form alloys with other metals | 21:26 |
genehacker | as I learned | 21:26 |
-!- cis-action [n=cis-acti@adsl-68-126-219-197.dsl.pltn13.pacbell.net] has quit [] | 21:31 | |
-!- nsh- is now known as nsh | 21:33 | |
kanzure | so I find a high quality paper (you can actually read it) and it's going over the genes necessary for natural competence | 21:52 |
kanzure | and the one diagram that I actually need to see, where the nucleotide sequence is given, | 21:52 |
kanzure | is static/noise. :( | 21:52 |
genehacker | damn college parties | 21:54 |
fenn | damn rogue CIA programs | 21:54 |
genehacker | I go to a college party, I think I am going to have a good time but no | 21:55 |
genehacker | they have REDACTED} | 21:55 |
genehacker | hmmm... | 21:56 |
genehacker | I wonder how hard it would be to make an IR spectroscopy unit | 21:57 |
kanzure | infrared LED + something something something | 21:58 |
kanzure | or see the cereal box + CD methods | 21:58 |
genehacker | I want to detect a certain chemical that isn't very complex in small concentrations in the air | 21:58 |
genehacker | so I can put it on a robot | 21:59 |
fenn | you want a mass spectrometer | 22:00 |
genehacker | yeah, like I can get a cheap one of those | 22:00 |
fenn | spectroscopy only works in high concentrations | 22:00 |
genehacker | sure they have some the size of game boys | 22:00 |
genehacker | I want to detect alcohol | 22:01 |
genehacker | ok? | 22:01 |
drazak | I can have things mass spec'd by the lady upstairs | 22:01 |
drazak | for probably free | 22:01 |
genehacker | I want to make a robot that goes around the the dorms automatically sampling the air and seeing if there is alcohol present | 22:02 |
genehacker | so that way people don't have parties with alcohol | 22:02 |
genehacker | and I can to parties without getting arrested | 22:02 |
nsh | wtf | 22:06 |
fenn | genehacker: that's the dumbest idea i've ever heard | 22:06 |
genehacker | of course, but people would still buy it | 22:07 |
fenn | hmm. an alcohol-sniffing robot could be useful, but not for your intended scenario | 22:07 |
nsh | have you been arrested for being at a college party with alcohol, genehacker? | 22:07 |
genehacker | DVD rewinders exist, and people buy them too | 22:07 |
genehacker | no | 22:07 |
nsh | then what's your point? | 22:08 |
fenn | actually it would probably sell better as a cellphone attachment | 22:08 |
genehacker | hmmm... | 22:08 |
genehacker | now all we have to do is figure out how to constuct a really really tiny alcohol sensor | 22:08 |
genehacker | detecting alcohol at low concentrations might cause some problesm | 22:10 |
genehacker | alcohol present in mouthwash or hand santizing wipes could set it off | 22:11 |
kanzure | ok. mission accomplished. | 22:12 |
genehacker | kanzure, could you tell me what parameters I need to feed that gear generator program of yours? | 22:12 |
genehacker | I know the gear needs to be able to withstand 500 oz/in of torque for sure | 22:12 |
kanzure | genehacker: unfortunately it's not my program. I don't actually have the source code laying around here. But IIRC it wants you to set some initial variables like xyz input location, and torque | 22:13 |
genehacker | it's a gear in a gear | 22:13 |
* fenn suspects someone is going to be disappointed | 22:15 | |
genehacker | I don't think that program can generate weird internal gears so | 22:16 |
genehacker | I'll just do it my self | 22:16 |
-!- jm|space [n=jm@p57B9D33D.dip.t-dialin.net] has quit [Read error: 113 (No route to host)] | 22:27 | |
-!- ybit [n=heath@unaffiliated/ybit] has quit [Remote closed the connection] | 22:38 | |
-!- nsh [n=nsh@host86-133-35-168.range86-133.btcentralplus.com] has quit [Read error: 110 (Connection timed out)] | 22:47 | |
-!- xp_prg [n=xp_prg3@208-106-52-115.adsl.dynamic.shocking.com] has quit ["This computer has gone to sleep"] | 23:22 | |
kanzure | so, sata wants me to design some sort of photovoltaic solar powered floatation device that flaps large-surface-area arms around wildly | 23:25 |
genehacker | hey I got a 10v solar cell you want it? | 23:26 |
genehacker | btw I'm taking dynamics I could do this | 23:26 |
kanzure | what does taking dynamics have to do with it | 23:27 |
genehacker | sounds like to me you want some sort of crank mechanism | 23:27 |
kanzure | well, for one, a mechanical system like this might not even be a good idea. | 23:27 |
kanzure | what happened to us thinking about jets? etc.? | 23:27 |
genehacker | heh | 23:27 |
genehacker | you should see what they use to aerate fish farms | 23:28 |
genehacker | they use paddle wheels that spin REALLY REALLY FAST | 23:28 |
fenn | what's wrong with a bubbler? | 23:28 |
fenn | super low power, high surface area, stirs the water around | 23:29 |
fenn | impossible to break | 23:29 |
genehacker | http://www.youtube.com/watch?v=meDcNK0-tio&feature=related | 23:29 |
* kanzure nods | 23:29 | |
fenn | if it does break, they cost so little as to be disposable | 23:29 |
kanzure | fenn: I am not convinced that this is going well at all | 23:29 |
kanzure | I've been in the group for how many months now? and have seen no efficiency charts, no nothin' | 23:30 |
kanzure | nor have I been able to create them for that matter | 23:30 |
kanzure | but that's another problem | 23:30 |
kanzure | I don't think 'efficiency' is a big issue there :/ | 23:30 |
kanzure | just throw money at it, that'll git'r'done! | 23:30 |
genehacker | http://www.youtube.com/watch?v=vDUyNom6hZ4&feature=PlayList&p=C1F9C0526125C311&index=0&playnext=1 | 23:30 |
genehacker | if it's solar-powered | 23:31 |
kanzure | where is boise? | 23:31 |
kanzure | hm, idaho. | 23:31 |
kanzure | is there some third world country named boise? because that's more likely to be what this person means | 23:31 |
kanzure | http://en.wikipedia.org/wiki/Boise_(disambiguation) | 23:31 |
kanzure | http://heybryan.org/~bbishop/docs/gears/CADgears2/modclasscorrect.bmp (just generated a few minutes ago) | 23:32 |
genehacker | I think something might be wrong | 23:33 |
genehacker | uhm seriously | 23:33 |
genehacker | did you look at the matlab script I gave you? | 23:33 |
fenn | why are you making .bmp files? | 23:33 |
genehacker | I mean it makes 3d models of gears | 23:34 |
kanzure | fenn: that's what pythonocc exports | 23:34 |
genehacker | sure, it's in easy matlab format | 23:34 |
kanzure | I would convert them, but that would be an extra step | 23:34 |
fenn | jesus that thing looks dangerous | 23:34 |
kanzure | ?> | 23:34 |
fenn | 'hey bubba lets weld some rotating spikes onto yet tractor' | 23:35 |
kanzure | heh | 23:35 |
genehacker | well, at least they thought to do aerating in the first place | 23:35 |
fenn | seriously, bubblers rock | 23:35 |
kanzure | fenn: if you want, you could come to the next lab meeting and knock some sense into everyone | 23:36 |
genehacker | hmmmm.... | 23:36 |
genehacker | so we need a solar powered air compressor? | 23:36 |
fenn | my fees are very reasonable | 23:36 |
kanzure | fenn: oh? | 23:36 |
genehacker | kanzure if you're gonna make it solar powered | 23:37 |
genehacker | look into making something beam style | 23:37 |
genehacker | BEAM style | 23:37 |
kanzure | beam robotics, or beam beams | 23:37 |
kanzure | okay | 23:37 |
fenn | beam is lame | 23:37 |
fenn | just charge a battery | 23:37 |
genehacker | do you call a giant walking robot with metal detectors on it that seeks out metallic objects lame? | 23:38 |
genehacker | (metallic objects largely being unexploded munitions) | 23:38 |
genehacker | the guy who built this giant robot once sent it on a path down a hall way to terrorize his secretary | 23:40 |
-!- ybit [n=heath@unaffiliated/ybit] has joined #hplusroadmap | 23:42 | |
-!- PeerInfinity [n=someone@216.36.180.162] has quit [] | 23:43 | |
fenn | hmm calculus. | 23:51 |
fenn | given a heatsink at 90C, mass 30g, heat capacity 0.9J/gK, and junction to ambient of 0.5 degree per watt, how long will it take to cool off to 40C, given an ambient temperature of 25C | 23:53 |
genehacker | oh god, not heat transfer | 23:54 |
genehacker | that class is hard | 23:54 |
* fenn looks at http://books.google.com/books?id=lIBCltUml6oC&pg=PA91&lpg=PA91&dq="long+will+it+take+to+cool+off"&source=bl&ots=RPhthHS2QE&sig=oyI3Rgj4V5elPbliUD6pShGKgtI&hl=en&ei=kK7qSaqdDOWwtgf1m4yXBg&sa=X&oi=book_result&ct=result&resnum=4 | 23:54 | |
fenn | i hate when a problem seems so simple but i can't figure out how to put it into equation form | 23:55 |
fenn | all these textbook problems are exactly the same | 23:56 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!